Acquired Amegakaryocytic Thrombocytopenia with Glycoprotein IIb/IIIa Antibody in a Patient with Mantle Cell Lymphoma

Abstract

Acquired amegakaryocytic thrombocytopenia (AAMT) is a rare disorder characterized by thrombocytopenia, marked megakaryocytic hypoplasia, and preserved other-lineage hematopoiesis in the bone marrow. The etiology of AAMT remains poorly understood owing to its rarity.

We encountered a diagnostically challenging case involving a 66-year-old man who showed severe thrombocytopenic bleeding with isolated megakaryocytic hypoplasia, elevated serum thrombopoietin levels, glycoprotein IIb/IIIa antibody positivity, and prolonged platelet transfusion refractoriness following mantle cell lymphoma (MCL). Treatment with corticosteroids and intravenous immunoglobulin was ineffective, while a combination of multiagent chemotherapy, including rituximab, was beneficial for both thrombocytopenia and MCL. Ultimately, the patient was diagnosed with AAMT and immune thrombocytopenia (ITP)-like platelet destruction.

This case suggests that AAMT and ITP are non-exclusive and sometimes overlap as components of a broad spectrum of platelet-related autoimmune diseases.

Introduction

Acquired amegakaryocytic thrombocytopenia (AAMT) is a rare disorder characterized by severe thrombocytopenia, with a marked reduction in megakaryocytes and preserved other-lineage hematopoiesis in the bone marrow (BM) (1,2). Previous reports indicate that AAMT is frequently misdiagnosed as immune thrombocytopenia (ITP), which is the most common autoimmune disease that leads to thrombocytopenia because AAMT is indistinguishable from ITP without a BM test (1,3-5). Treatments that are typically effective for ITP, such as corticosteroids or intravenous immunoglobulin (IVIG), are usually ineffective for AAMT, underscoring the clinical importance of distinguishing between ITP and AAMT (1,3). Although the etiology of AAMT remains poorly understood due to its rarity, it is often complicated by autoimmune diseases, such as systemic lupus erythematosus (SLE), suggesting an immune-mediated disease (1,6).

We encountered a case of severe thrombocytopenic bleeding with isolated megakaryocytic hypoplasia in the BM and elevated serum thrombopoietin (TPO) levels after the diagnosis of mantle cell lymphoma (MCL), suggesting AAMT. However, this case exhibited ITP-like findings, such as positivity for glycoprotein (GP) IIb/IIIa antibody and prolonged platelet transfusion refractoriness (PTR), complicating the diagnosis of thrombocytopenia.

Case Report

A 66-year-old man with a medical history of hypertension, cerebral infarction, and type 2 diabetes mellitus developed painless systemic lymphadenopathy, which had persisted for several months. A cervical lymph node biopsy revealed multiple follicle-like structures with monomorphic proliferation of medium-sized blastoid lymphocytes with prominent nucleoli and a high mitotic rate. On immunostaining, the atypical cells were positive for L-26, CD5, and CyclinD, and negative for CD10. Flow cytometry showed that the cells were positive for CD5, CD19, CD20, CD38, CD45, and kappa light chain restriction, and negative for CD10 and CD23. A G-band chromosomal analysis of the lymph node using G-banding was not possible due to the lack of an analyzable metaphase. However, fluorescence in situ hybridization (FISH) detected BCL1/IgH fusion signals in 70% of cells. BM aspiration revealed slightly hypoplastic fatty marrow with megakaryocytic hypoplasia and 1.2% abnormal lymphocytes, suggesting minor BM involvement of the MCL. A G-band chromosomal analysis of the BM revealed a normal karyotype, and the BCL1/IgH fusion signals by FISH were unevaluable. These findings led to a diagnosis of advanced blastoid MCL.

Fig. 1 shows the clinical course, with the first day of hospitalization set as Day 1. On a cervical lymph node biopsy (Day -24), the patient showed no bleeding tendency and the platelet count was within the normal range. However, 3 weeks later, the patient was admitted to our hospital because of lower-extremity petechiae and oral bleeding attributed to severe thrombocytopenia. As shown in Table 1, the laboratory data on admission revealed a WBC count and hemoglobin level within their normal ranges, but the platelet count dropped to 1.8×10⁴/μL. Since we did not have access to a software program for the analysis, the immature platelet fraction was not measured. No signs of disseminated intravascular coagulation (DIC). Platelet-associated immunoglobulin G (PAIgG) level was elevated. Whole-body computed tomography revealed systemic lymphadenopathy, without hepatosplenomegaly.

Figure 1.

Clinical course. IVIG: intravenous immunoglobulin, PSL: prednisolone, DEX: dexamethasone, CHOP: cyclophosphamide, adriamycin, vincristine, and prednisolone, EPOCH: etoposide, prednisolone, vincristine, cyclophosphamide, and adriamycin, FC: fludarabine and cyclophosphamide, TPO: thrombopoietin, MACE: modified antigen capture ELISA, PLT: platelet count, RET: reticulocyte count, FFP: fresh frozen plasma, RBC: red blood cell, PC: platelet concentrate

Table 1.

Laboratory Findings on Admission.

WBC	6,790	/μL		AST	16	IU/L		sIL-2R	2,600	U/mL
Neutro	68	%		ALT	10	IU/L		PAIgG	107	ng/107cells
Mono	6	%		LDH	196	IU/L		ANA	20	fold
Lym	19	%		ALP	304	IU/L		CRP	<0.01	mg/dL
RBC	420	×104/μL		T-bil	0.5	mg/dL		HBs-Ag	(-)
Hb	13.1	g/dL		TP	8.4	mg/dL		HCV-Ab	(-)
Ht	37.1	%		Alb	4.1	mg/dL		HIV-Ab	(-)
Plt	1.8	×104/µL		BUN	25.6	mg/dL		IgG	2,469	mg/dL
Ret	5.2	×104/µL		Cr	1.63	mg/dL		IgM	131	mg/dL
				Na	139	mEq/L		IgA	335	mg/dL
PT-INR	1.15			K	4.6	mEq/L
APTT	28.6	s
Fbg	328	mg/dL		anti-Helicobacter pylori IgG antibody					20	fold
FDP	3	μg/mL

WBC: white blood cell count, Neutro: neutrophil, Mono: monocyte, Lym: lymphocyte, RBC: red blood cell count, Hb: hemoglobin, PLT: platelet count, PT-INR: prothrombin time-international normalized ratio, APTT: activated partial thromboplastin time, Fbg: fibirinogen, FDP: fibrin degradation products, Alb: albumin, AST: aspartic transaminase, ALT: alanine transaminase, LDH: lactate dehydrogenase, ALP: alkaline phosphatase, T-bil: total bilirubin, TP: total protein, BUN: blood urea nitrogen, Cr: creatinine, sIL-2R: soluble interleukin-2 receptor, PAIgG: platelet-associated immunoglobulin, ANA: anti-nuclear antibody, HBs-Ag: hepatitis B surface antigen, HCV-Ab: hepatitis C antibody, HIV-Ab: human immunodeficiency virus antibody, CRP: C-reactive protein, IgG: immunoglobulin G, IgA: immunoglobulin A, IgM: immunoglobulin M

On Day 2, the patient's platelet count dropped to 0.3×10⁴/μL. BM re-aspiration revealed marked megakaryocytic hypoplasia with preserved erythroid and myeloid hematopoiesis (Fig. 2A). Additionally, 0.98% of lymphocytes in the BM were positive for CD5, CD19, CD20, and kappa light chain restriction on flow cytometric analysis (Fig. 2B), which is consistent with the typical phenotype of MCL. Although the presence of significant megakaryocytic hypoplasia appeared atypical of ITP, we tentatively diagnosed the patient with ITP associated with MCL. Treatment began with high-dose dexamethasone (40 mg/day) and IVIG, which did not improve the thrombocytopenia. The patient's platelet count remained lower than 0.2×10⁴/μL for 1 month, leading to prolonged severe melena. This condition necessitated daily transfusions of platelet concentrate (PC), red blood cells, and fresh frozen plasma to manage dilutional coagulopathy. During daily transfusions of PC from Days 2 to 32, no improvements in bleeding tendency or elevations in platelet count were observed. Notably, repeated analyses of 1-hour post-transfusion corrected platelet count increments (CCI-1h) of 10-20 units failed to demonstrate significant improvement, without any infectious diseases, DIC, fever, or splenomegaly. This finding suggests peripheral platelet destruction. On Day 7, anti-human leukocyte antigen antibodies were undetectable using a lymphocyte cytotoxicity test, anti-human immunoglobulin lymphocyte cytotoxicity test, and magnetic mixed passive hemagglutination method. However, anti-GP IIb/IIIa antibodies were detected using a modified antigen capture ELISA assay on the same day.

Figure 2.

Bone marrow findings. (A) On admission: Crush-spread preparation, ×200 magnification. Slightly hypoplastic marrow with megakaryocytic aplasia. Preserved erythroid and myeloid hematopoiesis, with minor involvement of mantle cell lymphoma (inset, ×1,000 magnification). (B) After chemotherapy plus rituximab (Day 46): Crush-spread preparation, ×200 magnification. Recovery of megakaryocytic hematopoiesis (red arrowheads) without any notable involvement of mantle cell lymphoma.

On Day 10, rituximab monotherapy was administered, which proved ineffective for both lymphadenopathy and thrombocytopenia. Given the refractoriness to treatment, we considered the possibility that thrombocytopenia was not solely caused by autoimmune consumption of platelets, but also by other etiologies, including impaired platelet production and a significant correlation with MCL. This prompted us to add antilymphoma chemotherapy to rituximab therapy. On Day 18, we administered cyclophosphamide, adriamycin, vincristine, and prednisolone (CHOP) multi-agent chemotherapy. On Day 21, the serum TPO level was notably high at 640 pg/mL, which was consistent with reduced megakaryocytes in the BM. Following rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone therapy, the patient's platelet count gradually improved, starting from 1.8×10⁴/μL on Day 35 and entering the normal range by Day 65, accompanied by a gradual reduction in systemic lymphadenopathy. A test for anti-GP IIb/IIIa antibodies was negative on Day 30. The last PC was transfused at a platelet count of 0.3×10⁴/μL on Day 32. BM aspiration on Day 46 showed recovery of megakaryocytic hematopoiesis with no MCL involvement on flow cytometry (Fig. 2). Unfortunately, we did not assess the PAIgG or TPO levels after R-CHOP therapy. Table 2 summarizes the chronological course, clinical status, platelet count, and BM findings, focusing on the megakaryocyte count and infiltrating lymphoma cells.

Table 2.

Chronological Course on Clinical Status, Platelet Count, and Bone Marrow Findings.

	Day -24	Day 2	Day 17	Day 46
Clinical status	Diagnosis of MCL	On admission	After Rituximab	After CHOP
Platelet count in peripheral blood (×104/μL)	21.2	0.3	0.1	11.5
Bone marrow findings
Nucleated cell count (/μL)	60,000	19,000	27,000	43,000
Total MgK count in all fields in each crush-spread preparation	39	3	0	13
CD5+, CD20+, κ-ch+cell on flow cytometry (%)	Not analyzed	0.98	0.95	0.0

MCL: mantle cell lymphoma, CHOP: cyclophosphamide, adriamycin, vincristine, and prednisolone, MgK: megakaryocyte

On Day 51, to intensify the treatment for MCL, we switched from R-CHOP to R-EPOCH therapy (rituximab, etoposide, prednisolone, vincristine, cyclophosphamide, and doxorubicin). However, after 2 cycles of R-EPOCH, the MCL progressed with the recurrence of asymptomatic mild thrombocytopenia within a platelet count range of 1.4-10.0×10⁴/μL. Interestingly, the CCI-24h of 10 units of PC for thrombocytopenia at this point was 3.2×10⁴/μL, showing improvement, unlike the initial phase. Several salvage chemotherapies were ineffective for mild thrombocytopenia or MCL. Ultimately, he died with leukemic progression (31% of MCL cells in peripheral blood) and pancytopenia (WBC 3,330/μL; hemoglobin 7.6 g/dL; platelet 1.7×10⁴/μL; reticulocyte count 4.4×10⁴/μL) 3 years after the initial diagnosis.

Discussion

We encountered a diagnostically challenging case of severe thrombocytopenic bleeding after the diagnosis of MCL. This thrombocytopenia did not respond to corticosteroids and IVIG, whereas multi-agent chemotherapy with rituximab was initially effective for both thrombocytopenia and MCL. Although isolated megakaryocytic hypoplasia in the BM and elevated serum TPO levels suggested AAMT, prolonged PTR and anti-GP IIb/IIIa antibody positivity suggested ITP, complicating the diagnosis of thrombocytopenia. Ultimately, we diagnosed the thrombocytopenia as AAMT with ITP-like platelet destruction.

Because AAMT can involve heterogeneous clinical conditions, its management can vary depending on the underlying cause. Previous reports have suggested that the etiology includes humoral autoimmunity against TPO-related antigens, T-cell autoimmunity against megakaryocytes, or exogenous pathogeneses involving some viruses, drugs, or malnutrition (2,4). Additionally, some cases previously classified as AAMT might actually represent early stages of BM failure, such as aplastic anemia (AA) which mimics AAMT (2,7). The majority of AAMT has been considered to be associated with autoimmune mechanisms (1). A report showed that more than half (67%) of AAMT cases had serum autoantibodies against the TPO receptor (anti-c-myeloproliferative leukemia protein antibodies), which was shown to be a type of functional autoantibody that inhibits megakaryocytic colony formation in vitro (8). Anti-c-Mpl antibodies were not measured in this case, because the test was unavailable at our institute.

The elevated serum TPO level in this case supported the diagnosis of AAMT. TPO, a key regulator of megakaryopoiesis and of other hematopoietic precursor cells, is predominantly produced in the liver. In the human body, TPO is regulated at a constant level by a balance between hepatic production and consumption by binding to c-Mpl expressed on megakaryocytes and hematopoietic stem cells (9). In AAMT and AA, serum TPO levels are elevated due to megakaryocytic hypoplasia. In contrast, in ITP, serum TPO levels are only slightly above normal. Therefore, measurement of the serum TPO level is clinically useful for distinguishing AAMT and AA from ITP (1). In this case, although the BM showed hypoplastic fatty marrow in addition to megakaryocytic hypoplasia, the peripheral blood showed white blood cell counts within the normal range and elevated reticulocyte counts in response to severe bleeding, suggesting the presence of AAMT rather than AA.

In the present case, anti-GP IIb/IIIa antibodies were transiently detected during severe thrombocytopenia. This is the most frequent autoantibody in ITP cases and shows high diagnostic specificity (10). Moreover, the patient exhibited prolonged PTR and reduced CCI-1h. Although there is still room for debate, there are reports that PTR without CCI-1h improvements suggests immune consumption (11). Therefore, our case of AAMT may involve ITP-like platelet destruction.

The relationship between AAMT and ITP is non-exclusive and sometimes overlapping. A previous report showed that anti-c-Mpl antibodies were positive in 22% of patients with SLE-associated thrombocytopenia with megakaryocytic hypoplasia and in 10% of patients with ITP; the majority of these patients also had concomitant anti-GP IIb/IIIa antibodies (6,10). Indeed, in previous case reports on AAMT, a combination of megakaryocytic aplasia and ITP-like platelet destruction was observed (12-14). Previous reports have shown that, in ITP cases, antiplatelet antibodies or autoreactive T-cells inhibit megakaryocytic maturation, impair platelet production, and promote platelet destruction via diverse immune networks involving B-cells, T-cells, and the reticuloendothelial system (15,16). Currently, a comprehensive concept is emerging in which ITP-type platelet destruction targeting GP IIb/IIIa and AAMT-specific impaired megakaryocytic functions targeting c-Mpl are considered to overlap as part of a spectrum of platelet-related autoimmune diseases (1). Considering this inclusive concept, the ITP-like platelet destruction in this case was consistent with the diagnosis of AAMT.

In lymphoma-associated ITP, especially non-Hodgkin's lymphoma, reviews of case reports have shown that corticosteroids and IVIG are typically ineffective, and rapid anti-lymphoma chemotherapy is the optimal treatment strategy (17,18). To our knowledge, reports on chemotherapy for lymphoma-associated AAMT are limited to 3 cases: a pediatric Hodgkin's lymphoma case showing improved thrombocytopenia 4 months after the initiation of multiagent chemotherapy (19); an elderly B-cell lymphoma case exhibiting transient, slight improvement in thrombocytopenia 2 weeks after the administration of vinblastine (20); and our MCL case. Multiagent chemotherapy has the potential to improve the platelet-related autoimmune network, including AAMT and ITP, alongside direct targeting of lymphoma. This highlights the importance of rapid chemotherapy as a comprehensive strategy for the management of lymphoma-associated thrombocytopenia.

A major diagnostic limitation of this case was the lack of a BM biopsy test. It is critically important to precisely assess megakaryocyte cellularity using a BM biopsy combined with a test for anti-CD42b antibodies, which is specific for megakaryocytes, when diagnosing AAMT. Unfortunately, a BM biopsy was avoided because of the bleeding tendency. The BM aspirate did not solidify, and immunochemical evaluation using anti-CD42b antibodies could not be performed. A previous study suggested that the crush-spread technique is more valuable than the wedge-spread technique for evaluating megakaryopoiesis in the BM (21). Therefore, we determined the total megakaryocyte count in all fields in each crush-spread preparation as a substitute for the unperformed BM biopsy. As shown in Table 2, a decreased megakaryocyte count using this method seemed to be associated with the severity of thrombocytopenia.

In this case, the potential link between minor BM involvement of the MCL and AAMT was notable. At the onset of AAMT, the occupation rate of the MCL in the BM was approximately 1%, which generally did not cause BM suppression due to the tumor burden. After the disappearance of MCL in the BM due to R-CHOP therapy, megakaryocytic hypoplasia and thrombocytopenia were transiently resolved, suggesting an association between minor BM involvement and AAMT. However, after the relapse of MCL involving the BM, mild thrombocytopenia occurred, although severe thrombocytopenia was not observed. Interestingly, elevated CCI-24h of 10 units of PC in this relapse phase suggested an improvement in platelet destruction. Although not performed, chronological observation of the serum TPO levels and anti-GP IIb/IIIa antibodies might have helped to elucidate the mechanism of later-phase thrombocytopenia.

In conclusion, we encountered a diagnostically challenging case of severe bleeding from AAMT with ITP-like platelet destruction after the diagnosis of MCL, which was successfully managed with multiagent chemotherapy including rituximab until the progression of MCL. This suggests an overlapping association between AAMT and ITP as components of a broader spectrum of platelet-related autoimmune diseases.

The authors state that they have no Conflict of Interest (COI).